Full Terms & Conditions of access and use can be found athttp://www.tandfonline.com/action/journalInformation?journalCode=talc20

Alcheringa: An Australasian Journal of Palaeontology

ISSN: 0311-5518 (Print) 1752-0754 (Online) Journal homepage: http://www.tandfonline.com/loi/talc20

Leaf preservation in Eucalyptus woodland as amodel for sclerophyll fossil floras

Gregory J. Retallack

To cite this article: Gregory J. Retallack (2018): Leaf preservation in Eucalyptus woodland asa model for sclerophyll fossil floras, Alcheringa: An Australasian Journal of Palaeontology, DOI:10.1080/03115518.2018.1457180

To link to this article: https://doi.org/10.1080/03115518.2018.1457180

Published online: 06 May 2018.

Submit your article to this journal

View related articles

View Crossmark data

Leaf preservation in Eucalyptus woodland as a model forsclerophyll fossil floras

GREGORY J. RETALLACK

RETALLACK, G.J., May 2018. Leaf preservation in Eucalyptus woodland as a model for sclerophyll fossil floras. Alcheringa xxx, xxx–xxx.

A comparison of 29 identifiable vascular plant species in litter beneath Eucalyptus woodland with at least 74 species living nearby showed that thelitter is a poor representation of standing vegetation. The leaf litter is dominated by sclerophyll leaves, which are a factor of 6.2 over-representedin litter for Angophora costata, factor of 5.7 for Melaleuca linariifolia, of 3.6 for Eucalyptus spp., of 3.5 for Pteridium esculentum and of 2.1 forAcacia linifolia. Angophora leaves are favored by lignification, with denser venation than Eucalyptus leaves. Sparse emergent oil glands of Ango-phora also provide fewer entry points for bacteria than rotted internal oil glands of Eucalyptus. The myrtaceous taxa Angophora, Eucalyptus, Mela-leuca and Kunzea all have oils dominantly of preservative terpene. Melaleuca linariifolia and Acacia linifolia also have leaves and phyllodes(respectively) that are narrow with a thick lignin midrib. Thickly cuticled, succulent, hirsute, pubescent, and pinnate leaves, and green stems arenot favored for preservation, because they rot from the inside out. Conspicuously absent in the leaf litter are nonsclerophyll leaves, most grassesand low herbs. This modern sclerophyll leaf litter matches Sydney Basin Permian and Triassic fossil plant localities above nutrient-poor siliceouspaleosols, which may have had much more diversity than the preserved fossil flora. Clayey calcareous paleosol leaf litters and lake deposits mayrecord a truer record of local floristic diversity in deep time than sclerophyll leaf litters.

Gregory J. Retallack [gregr@uoregon.edu], Department of Earth Sciences, University of Oregon, Eugene, OR 97403-1272, USA. Received27.11.2017; revised 24.1.2018; accepted 5.3.2018.

Key words: Taphonomy, leaf litter, Angophora, Dicroidium, Lepidopteris.

PLANT TAPHONOMY has emphasized physical sedi-mentary comminution and distribution of fossil leaves(Ferguson 1985, Spicer 1991, Gastaldo et al. 2005), butanother filter for what remains to join the rock record isdecay within leaf litters of soils (Burnham 1989, Green-wood 1992). Many fossil leaf deposits and plant tapho-nomic studies have been based on lacustrine (Drake &Burrows 1980, Ferguson 1985, Hill & Gibson 1986,Gastaldo et al. 1989, Spicer 1991, Sniderman et al.2013, Astorga et al. 2016), deltaic (Gastaldo 1989, Gas-taldo et al. 2005), or fluvial depositional models(Howarth & Fisher 1976, Blackburn & Petr 1979, Steartet al. 2002), but there also are fossil leaf litters, markedby root traces, small plants in growth position, and mat-ted, skeletonized, and dry-curled leaves (Retallack1977a, Retallack et al. 2000, Retallack & Dilcher 1988,2012). Fossil leaf litters are much more common thantypically appreciated, like associated paleosols, whichhave only been generally recognized in the past fewdecades (Retallack 2013a). Table 1 enumerates 102 leaflitter localities represented by plant fossils in the Con-don Collection of the Museum of Natural and CulturalHistory of the University of Oregon (online portal:paleo.uoregon.edu). Leaf litter localities are 14.6% ofthe total of 697 plant localities in the Condon Collec-

tion, which has 35.0% lacustrine plant localities, 16.0%permineralized stumps rooted in paleosols, 16.0% local-ities in coal-roof clays, 10.9% plant localities in marinerocks, 4.9% localities within coal underclays, and 2.4%coal ball localities. These other kinds of deposits weredescribed by (Scott 1977a, 1977b, Retallack 1981,1997a, Retallack & Dilcher 1988, Gastaldo et al. 1995,Scott et al. 1996). Fossil leaf litter assemblages arewidely recognized as ‘foliar roofs’ (Krassilov 1975,Holmes 1982), and as ‘obrution deposits’ (Libertínet al. 2009, Stevens & Hilton 2009, Dunn et al. 2012).Fossil leaf litters have also been called ‘silcrete plantfossils’ (Rozefelds 1990, Carpenter et al. 2011), ‘ganis-ters’ (Retallack 1977a, Percival 1983), and ‘nut beds’(Manchester 1994, Retallack et al. 2000). Like permin-eralized fossil forests, coal balls, underclays, and coal-roof floras, fossil leaf litters are a record of vegetationin its place of growth within a sedimentary depositionalbasin (Retallack 1977a, Steart et al. 2006), unlikeleaves in lake or marine deposits mixed from severalsurrounding communities (Retallack 1985, Retallacket al. 2000).

Fossil leaf litters have their own biases as records ofpast vegetation because of decay: extremes of preserva-tional quality are mull and mor humus. Finely commin-uted mull humus of grassland soils (Mollisols) retainslittle identifiable plant material (Barratt 1968, Sanborn& Pawluk 1989). Plant debris is also actively commin-© 2018 Geological Society of Australia Inc., Australasian Palaeontolo-

gists https://doi.org/10.1080/03115518.2018.1457180

Published online 06 May 2018

Ma Age Formation Location Locality no.

4 Pliocene Deschutes Formation Madras, Oregon 121575 Pliocene Yonna Formation Klamath Falls, Oregon 157865 Pliocene Troutdale Formation Troutdale, Oregon 1227412 Miocene Mehrten Formation Gold Lake, California 1256014 Miocene Fort Ternan Beds Fort Ternan, Kenya 1167516 Miocene Cucaracha Formation Gaillard Cut, Panama 1242316 Miocene Imnaha Basalt Juliaetta, Idaho 1349017 Miocene Little Bay Shale Maroubra, New South Wales 1348917 Miocene Teanaway Basalt Teanaway, Washington 1311918 Miocene Kaswanga Tuff Rusinga Island, Kenya 1140618 Miocene Deschutes Formation Gateway, Oregon 1172021 Miocene Eagle Creek Formation Stevenson, Washington 1322321 Miocene Eagle Creek Formation Eagle Creek, Oregon 1348723 Miocene Sardine Formation Collawash, Oregon 1068823 Miocene Little Butte Volcanics Disston, Oregon 1150225 Oligocene Little Butte Volcanics Sweet Home, Oregon 1128330 Oligocene Little Butte Volcanics Springfield, Oregon 1223530 Oligocene Yaquina Formation Seal Rock, Oregon 1254732 Oligocene John Day Formation Painted Hills, Oregon 1187135 Eocene Ione Formation La Porte, California 1256135 Eocene Fisher Formation Goshen, Oregon 1142937 Eocene Fisher Formation Comstock, Oregon 1122638 Eocene Swauk Formation Liberty, Washington 1096843 Eocene Clarno Formation Sheep Smother Spring, Oregon 1215444 Eocene Clarno Formation Clarno, Oregon 1173345 Eocene Tukwila Formation Seattle, Washington 1084250 Eocene silcrete Bevendale, New South Wales 1269751 Eocene Bridger Formation Blue Rim, Wyoming 1251454 Paleocene Hanna Formation Hanna, Wyoming 1577755 Paleocene Herren Formation Arbuckle Mountain, Oregon 1164856 Paleocene Herren Formation Denning Spring, Oregon 1214956 Paleocene Eyre Formation Stuart Creek, South Australia 1249957 Paleocene Chuckanut Formation Bellingham, Washington 1221559 Paleocene Dawson Arkose Castle Rock, Colorado 1326266 Cretaceous Hell Creek Formation Buffalo, North Dakota 1190575 Cretaceous Laramie Formation Colorado Springs, Colorado 1326095 Cretaceous Windrow Formation Springfield, Minnesota 1068796 Cretaceous Dakota Formation Hoisington, Kansas 1166499 Cretaceous Dakota Formation Kanapolis, Kansas 11372130 Cretaceous Days Creek Formation O’Brien, Oregon 13488150 Jurassic Riddle Formation Thompson Creek, Oregon 13365168 Jurassic Curio Bay Beds Curio Bay, New Zealand 12395168 Jurassic Coon Hollow Formation Pittsburgh Landing, Idaho 12487170 Jurassic Cloughton Formation Cayton Bay, Yorkshire 10832175 Jurassic Marburg Subgroup Durikai, Queensland 11172210 Triassic Pekin Formation Gulf, North Carolina 13160229 Triassic Stockton Formation Phoenixville, Pennsylvania 10913230 Triassic Cacheuta Formation Potrerillos, Argentina 13462230 Triassic Molteno Formation Sterkstroom, South Africa 10779236 Triassic Falla Formation Schroeder Hill, Antarctica 12013240 Triassic Tank Gully Coal Measures Tank Gully, New Zealand 11984241 Triassic Fremouw Formation Fremouw Peak, Antarctica 13269242 Triassic Nymboida Coal Measures Nymboida, Australia 10611245 Triassic Bringelly Shale Camden, New South Wales 12941247 Triassic Lashly Formation Allan Hills, Antarctica 11935248 Triassic Camden Head Claystone Camden Head, New South Wales 11168248 Triassic Newport Formation Avalon, Australia 10364248 Triassic Fremouw Formation Graphite Peak, Antarctica 12033252 Triassic Weller Coal Measures Allan Hills, Antarctica 11943253 Permian Coal Cliff Sandstone Oakdale Colliery, N.S.W 12133253 Permian Buckley Formation Graphite Peak, Antarctica 12031256 Permian Newcastle Coal Measures Swansea, Australia 11174262 Permian Weller Coal Measures Portal Mountain, Antarctica 12405263 Permian Longtan Formation Meishan, China 12087

(Continued)

2 G. J. RETALLACK ALCHERINGA

uted in tropical soils (oxisols) by armies of termites andleaf-cutter ants (Lavelle et al. 1993). At the otherextreme is mor humus of conifer soils (Spodosols), con-sisting of thick accumulations of well-preserved, pineneedles (Tian et al. 1997). Between these extremes isthe mix of well-preserved and partially decayed remainsof the leaf litter type known as moder (Ponge 2003), asstudied here. Leaf litters provide easily accessible exam-ples of decay of leaves from the inside out, with nutri-tious mesophyll attacked by bacteria and fungi beforerefractory cuticle, and finally lignin of veins remainingas a leaf skeleton (Spicer 1991, Tian et al. 1997). Thus,particular kinds of leaves last longer in litter, and leafquality measures, such as carbon/nitrogen, and lignin/ni-trogen ratios, may be predictors of rates of leaf decom-position (Hannon 1956, 1958, Berg et al. 1996, Berg &Matzner 1997, Aerts 1997, Berg 2000). Past studies ofleaf litter representation of vegetation (Burnham 1989,Greenwood 1992, Ellis & Johnson 2013) have beenundertaken in young, often disturbed, fertile landscapes

(YODFEL of Hopper 2009), but this study examines anold climatically buffered infertile landscape (OCBIL ofHopper 2009). This study of a modern leaf litter aimsto document the features of leaves that are resistant todecay in an oligotrophic sclerophyll woodland as amodern analog for fossil leaf litters of oligotrophic pale-osols in deep time.

Location of studyThe study area is a grid laid out on a track 200 m froman origin at S33.764347° E151.109374°, west of Cullo-den Road and 50 m north of Motorway M2, north ofMacquarie University in the northern suburbs of theSydney metropolitan area, NSW (Fig. 1). The track ison a plateau and the sampled grid slopes from theredown 20 m in elevation toward Mars Creek, some100 m south of its junction with the Lane Cove Riverat Brown’s Waterhole (Fig. 2).

Table 1. (Continued).

Ma Age Formation Location Locality no.

265 Permian Vyrheid Formation Vereeniging, South Africa 15749270 Permian Kazankov-Martin Formation Novokutznesk, Siberia 10808289 Permian Vale Formation Lake Abilene, Texas 11671300 Pennsylvanian Pituil Formation Barreal Hill, Argentina 13455303 Pennsylvanian Alykaev Formation Novokutznesk, Siberia 10800305 Pennsylvanian Organ Rock Shale Moab, Utah 10857306 Pennsylvanian Mazurov Formation Novokutznesk, Siberia 10794310 Pennsylvanian Seaham Formation Lochinvar, New South Wales 10275319 Pennsylvanian Sheffield Blue Ganister Langsett, England 11409320 Pennsylvanian Glen Eyrie Formation Manitou Springs, Colorado 10896321 Pennsylvanian Spotted Ridge Formation Paulina, Oregon 11357322 Pennsylvanian Pocahontas Formation Ghent, West Virginia 11764345 Mississippian Mauch Chunk Formation Scranton, Pennsylvania 12537347 Mississippian Calciferous Sandstone Oxroad Bay, Scotland 10949348 Mississippian Calciferous Sandstone Foulden, Berwickshire 10834361 Devonian Wutong Formation Kongshang, China 12076361 Devonian Hampshire Formation Valley Head, West Virginia 11147362 Devonian Duncannon Member Burtville, Pennsylvania 11330363 Devonian Duncannon Member Hyner, Pennsylvania 12333370 Devonian Witpoort Formation Grahamstown, South Africa 15750374 Devonian Oneonta Formation Prattsville, New York 11610374 Devonian Oneonta Formation Pond Eddy, New York 11605375 Devonian Mandagery Formation Bindogandri Creek, N.S.W. 12699380 Devonian Walton Formation Hancock, New York 10866383 Devonian Oneonta Formation East Windham, New York 12003384 Devonian Oneonta Formation West Durham, New York 12507395 Devonian Wojciechowice Formation Zachełmie, Poland 13096400 Devonian Nellenköpfschichten Alken-an-der-Mösel, Germany 13070403 Devonian Stadfeldschichten Müsch, Germany 12124404 Devonian Hünsruckschiefer Waxweiler, Germany 12120405 Devonian Beacon Heights Orthoquartzite West Beacon, Antarctica 11956425 Silurian Bloomsburg Formation Palmerton, Pennsylvania 12822438 Silurian Shawangunk Formation Delaware Water Gap, New Jersey 12828443 Ordovician Juniata Formation William Bean Gap, Tennessee 13203444 Ordovician Juniata Formation Potters Mills, Pennsylvania 12332464 Ordovician Douglas Dam Member Douglas Dam. Tennessee 15725484 Ordovician Grindstones Range Sandstone Grindtone Range, South Australia 12378

Table 1. Fossil leaf litter localities in the Condon Collection (online portal: paleo.uoregon.edu).

ALCHERINGA LEAF LITTER PRESERVATION 3

Bedrock is Middle Triassic (Anisian) HawkesburySandstone, which crops out as low benches in the sam-pled slope (Mayne et al. 1974). The uppermost samplelocations along the track have 1 m of weathered sand-stone and shale of the Middle Triassic (Ladinian) Mit-tagong Formation (Retallack et al. 2011). Soils areimmature, sandy, quartz-rich Tenosols with low basesaturation and acid reaction, and limited supply of phos-phorus, nitrogen and potassium (Hannon 1956, Beadle1962, 1968, Chapman & Murphy 1989, Chittleborough1991, McKenzie et al. 2004). The climate from 1970 to2016 has been warm temperate, humid (AustralianBureau of Meteorology 2016), with mean annual pre-cipitation of 1156 mm and mean annual temperature of17.0 °C at Macquarie Park, 2 km south (Fig. 3). A dis-tinctive feature of this and nearby weather stations is aJune peak in winter precipitation.

Vegetation is woodland of scribbly gum, Eucalyptushaemastoma (Beadle, 1981), but there are variationsbetween this plateau sclerophyll woodland with 6 mcanopy and woodland of Mars Creek with 8 m canopy(Lake & Leishman 2004). On the rocky break in slope,Eucalyptus haemastoma is less common than E. piper-ita (Sydney peppermint) and Angophora costata (rosegum). The flora of the Hawkesbury Sandstone plateausaround Sydney is famous for its shrub diversity in Pro-teaceae, Fabaceae and Ericaceae (Beadle 1981). It is an

Fig. 1. Map of study area and experimental design east of Mars Creeknorth of Macquarie, University, North Ryde, N.S.W.

Fig. 2. Eucalyptus woodland near Browns Waterhole, on Lane Cove River, near junction with Mars Creek.

4 G. J. RETALLACK ALCHERINGA

OCBIL with high diversity but low fertility similar tothe kwongan of Western Australia, fynbos of SouthAfrica, and campo rupestre of Brazil (Hopper 2009, Sil-veira et al. 2016). Valley-bottom trees include nonscle-rophyll native plants, such as black wattle (Callicomaserratifolia) and coachwood (Ceratopetalum apetalum).Introduced trees (Ligustrum sinense), climbers (Lon-icera japonica), grasses (Cenchrus clandestinus) andgotu kola (Centella asiatica) were very common atBrowns Waterhole in the Lane Cove River, a disturbedaccess point for plant invasion just beyond the studyarea.

Materials and MethodsSite sampling was done on compass traverse from abaseline anchored at S33.764347° E151.109374° north-east of Macquarie University and oriented northeast(46° magnetic azimuth) along a track, with orthogonaltraverses down to Mars Creek to the west. Litter wassampled at 10 sites on June 16, 1972 and located usinga random number table from the grid (Fig. 1). This wasdone to approximate a typical collecting area and ran-dom outcrop of fossil plant localities. Litter was takenfrom a square 20 × 20 cm all the way down (15–20 cm) into leafless soil: for a volume of 0.8 m3 andweight of 1–2 kg, including soil discarded when leavesand other plant debris were cleaned. The total volumeof litter analyzed in all ten quadrats was 8 m3 and15.7 kg. Each litter sample was designed to mimic asmall fossil pit though a paleosol A horizon, andyielded more leaves than the 0.5 m2 quadrat used inplant paleoecology (Scott 1977a, 1977b, Scott & Collin-son 1983). The mean and standard deviation for theidentifiable items per quadrat were 16.3 ± 1.7 and ofthe species per quadrat were 9.0 ± 2.5. The sampleswere air-dried for a week, then sorted into species andweighed. Weight was used because it was easier tomeasure than leaf area commonly used in plant paleoe-cology (slab cover of Scott 1977a, 1977b, Scott & Col-linson 1983) and quadrat cover of plant ecology (Curtisand MacIntosh 1951, Méndez-Toribio et al. 2014). All

16 species found in the leaf litter show a good correla-tion between leaf area and weight (Fig. 4). Leaves ofAngophora costata 13 cm long by 3 cm wide havemean areas of about 1950 mm2 and mean weights of2.4 g, whereas less lignified 2.2 g leaves of Eucalytpushaemastoma 15 cm long by 3 cm wide have areas ofabout 2250 mm2. These dominant plants of the area arethus mesophyll in the scoring system of Wolfe (1993).Numbers of countable parts were also recorded,together with the number of quadrats with a record ofthat species (frequency). Frequency, number of partsand dry weight were combined into a litter relativeimportance value (LRIV) combined from each of threerelative importance values for each species, defined asfollows.

LRIV = frequency (% quadrats with species/totalquadrats) + items (% pieces of species/total pieces) +dry weight (% grams of species/total grams).

This adapts an approach used widely in plant ecol-ogy (Curtis & MacIntosh 1951, Méndez-Toribio et al.2014), known as importance value (IV, Rodrigues et al.2004) or species importance value (Kanade et al. 2008).Paleoecology of fossil plants commonly uses only onemeasure such as ‘slab cover’ (Scott 1977a, 1977b, Scott& Collinson 1983), which can be gained from correla-tion with weight (Fig. 4), but importance values addother dimensions for evaluating species representation.

The vegetation survey aimed to capture changes invegetation down the slope, in June 1972. Quadrats25 × 100 cm (0.25 m2) in size were placed every 4 mdown the slope for a total of 50 quadrats in three tran-sects 20 m apart on the west-facing slope only, becausethe litter samples fell on that slope. This quadrat sizewas selected because the scale of interest was leaves0.5–16 cm long. Cover was estimated on a 6-pointscale: 1 = 1–5%, 2 = 5–25%, 3 = 25–50%, 4 = 50–75%,5 = 75–95%, 6 = 95–100%. Frequency (number ofquadrats with that species) and density (number of indi-viduals of a species overhanging each quadrat) werebased on main trunks including those rooted outside thequadrat, but on tufts in the case of bunch grasses. Themean number of species and standard deviation perquadrat was 7.2 ± 0.39, which is 6% of the mean. Fre-

Fig. 3. Climate of Macquarie Park (Australian Bureau of Meteorology2016).

Fig. 4. Relationship between leaf area and weight in species found inleaf litter of Mars Creek, N.S.W. Many leaves in this vegetation aresmall and light weight.

ALCHERINGA LEAF LITTER PRESERVATION 5

quency, density and cover were added to make a vege-tation relative importance value (VRIV), as follows:

VRIV = frequency (% quadrats with species/total quadrats)+ density (% numbers of species per quadrat/total number for species) + cover (average% light interception of species /quadrat)

Representation of particular species in litter com-pared with standing vegetation was calculated as apreservability index (PI), based on the ratio of LRIVand VRIV as follows:

PI ¼ LRIV þ 1

VRIV þ 1

The constant of 1 was added to both denominatorand numerator to correct for anomalous values of infin-ity when the numerator was zero. When the propor-tional representation of species in both litter andvegetation is equal, the PI is 1. Leaves that are particu-larly decay-resistant have a PI greater than 1.

The approach used here is novel compared withother studies of leaf litter taphonomy, which have useda census approach to modern vegetation and a leaf areaapproach to litter (Burnham 1989, Greenwood 1992,Ellis & Johnson 2013). Census data do not take intoaccount the number of leaves produced by each individ-ual plant, which is difficult to assess independently(Ellis & Johnson 2013). The approach used hereincludes leaf production by comparing both dead andliving leaves on comparable scales (quadrat), and onmultiple measures (weight-area, frequency, number),because this is the scale of fossil leaf litter assemblagesatop paleosols (Retallack 1977a, Retallack & Dilcher1988, 2012, Retallack et al. 2000). The additional mea-sures of frequency and number are also useful for Per-mian leaf litters of the Coalcliff Sandstone and Triassicleaf litters of the Newport Formation of New SouthWales (Retallack 1997a, 1999, Retallack et al. 2011),chosen for comparison with this study.

ResultsLRIVs and their component metrics are shown in Fig. 5for all the species and various parts of plants recog-nized. These values and the PI for particular species arecompared in Fig. 6. Only seven out of 74 species aremore important in litter than they are in vegetation (PI> 1). This dominance of only a few species is reflectedin a broad plateau on a rarefaction curve of number ofspecies with continued sampling (Fig. 7).

Vegetation survey

The vegetation survey revealed 74 species in this smallarea (2340 m2 or 0.23 ha) of woodland (Fig. 6). Somegrasses were common identifiable species, but wintersampling did not allow identification of six distinctivegrass leaf forms. Lichens were not identified because

they are rarely encountered in fossil plant assemblages,and were mostly crustose forms on rocks. Only one

stand was dominated by bracken (Pteridium esculen-tum), and it was close to a litter sample. This surveywas especially useful in identifying a diversity of nativesedges (Chordifex fastigiatus and Carex polyantha inorder of importance) and sclerophyllous shrubs (Acacialongifolia, Lasiopetalum ferrugineum, Banskia serrata,Grevillea buxifolia and Hakea sericea, in order ofimportance). The relative order of importance of treeswas Eucalyptus piperita (13.4 VRIV), Melaleuca linari-ifolia (13.1), Eucalyptus haemastoma (4.1), Angophorahispida (3.3) and Angophora costata (3.1).

Leaf litter survey

Only 29 species were recognized in the 10 leaf litter sam-ples, and many of them were represented by both repro-ductive material and leaves (Fig. 5). Reproductivestructures were much rarer than leaves, as is typical offossil leaf litters (Retallack 1977a, Retallack et al. 2000).Some of the herbaceous taxa, such as Dianella (Aspho-delaceae) and grasses were represented by both live(green and pliable) and dead (dry and brown) material inthe litter samples. Two species of sedge (Chordifex fasti-giatus and Baloskion tetraphyllum) and of Eucalyptus (E.haemastoma and E. piperita) could be distinguishedwhen complete, but not from fragments; and fragmentswere combined in the litter metrics (Fig. 6).

Individual litter sample sites are very uneven in spe-cies representation, with many species recorded in onlyone quadrat. Most species recorded in only one sampleare rare in that sample, but bracken fern (Pteridiumesculentum) was an exception, locally dominating onelitter sample.

Twigs and other unidentifiable debris were most com-mon in the litter samples, followed by leaves; and repro-ductive structures were less common (Fig. 5). The mostimportant leaves in the litter were Melaleuca linariifolia(74.9 LRIV), Eucalyptus spp. (24.5), Angophora costata(12.7) and, to a lesser extent, Acacia linifolia (10.5). Bydry weight alone, the most important leaves in the 10 lit-ter samples were Eucalyptus spp. (12.4%), Angophoracostata (4.5%) and Pteridium esculentum (8.6%). Thehigh relative importance of M. linariifolia and A. linifoliais due to the large numbers and frequency of these smalllinear leaves in the litter samples.

PI (Preservability Index)

PIs show that leaves of Angophora costata werefavored for preservation over other leaves by a factor of

6 G. J. RETALLACK ALCHERINGA

6.2. Other leaves in the litter in order of PI were Mela-leuca linariifolia (PI 5.7), Eucalyptus spp. (3.6), Pterid-ium esculentum (3.5), Acacia linifolia (2.1), Alectryonsubcinereus (2.7) and Kunzea ambigua (2.3). The linearleaves of Melaleuca, Acacia and Kunzea attain highrepresentation by number of parts rather than by dryweight. The fern Pteridium is an exceptional case,which dominated only one sample. Native quince (Alec-tryon subcinereus, Sapindaceae) is the only other softand pliable, nonsclerophyll leaf well preserved in thelitter. The other 66 species known in the local vegeta-

tion have a PI of 1 or less, and so less than even oddsof preservation.

DiscussionHumification processes

Changes to plant material on the ground begin withphysical destruction by wind, rain, hail and sand abra-sion (Ferguson 1985). Waves of bacteria turn the leavesblack to brown with oxidizing enzymes, and some

Fig. 5. Relative importance index of identifiable objects in the leaf litter east of Mars Creek, N.S.W.

ALCHERINGA LEAF LITTER PRESERVATION 7

leaves then can bleach with age (Spicer 1991). Litterwith mainly bleached and intact leaves is called morhumus (Tian et al. 1997). Fungi and cyanobacteriacause leaves to skeletonize and aggregate, stucktogether with slime, in a humus type called moder(Ponge 2003). Mites and springtails, insect larvae andearthworms eat the leaves and excrete a finely commin-uted mull humus (Barratt 1968, Sanborn & Pawluk

1989). The litter and humus of Mars Creek is mor tomoder from top to bottom, with many intact and someskeletonized leaves, but little mull. This is because mostof the species are scleromorph, with thick cuticle, hypo-dermis, lignified and tannin-bearing cells and sunkenstomata. The scleromorphy in this case is not inter-preted as xeromorphy owing to inadequate moisture,but rather peinomorphy owing to low soil nutrients

Fig. 6. Preservability index for species in leaf litter compared with standing vegetation east of Mars Creek, N.S.W.

8 G. J. RETALLACK ALCHERINGA

(Retallack 2009). Hannon (1956) showed that there isadequate precipitation and soil moisture, but little nitro-gen in the soil or parent rock. Beadle (1962, 1968)extended this work to show that phosphate and potas-sium is also limiting for Hawkesbury Sandstone plateaufloras.

Because of this slow decomposition in the leaf litterof Mars Creek, dry leaves accumulate until theybecome a fire hazard, as in other eastern Australian for-ests (Fox et al. 1979). The last ground fire through thestudy area was in 1967, before sampling in 1972. Char-coal fragments in the litter samples were included in thecategory of twigs (Fig. 5). Thus, the litter samples rep-resent accumulation over five years.

Summer rain near Sydney is often in the form ofviolent, late afternoon thunderstorms, which wash awayleaves and pile them into litter dams (Mitchell & Hum-phreys 1987). These thick piles decay while moist untilthe litter dries, so that there is small-scale lateral hetero-geneity in quality of humus.

Leaf features preferentially preserved

The dominance of Angophora costata over Eucalyptusin leaf litter was also noted by Hannon (1956), and isespecially striking considering that rose gum is not alarge proportion of the standing vegetation (Fig. 6).Angophora costata leaves have ‘extremely dense vena-tion’ (0.21 mm tertiary vein spacing) comparable withA. hispida (0.22 mm), unlike ‘sparse reticulation’ ofEucalyptus haemastoma (1.7 mm) and E. piperita(0.45 mm: Brooker & Nicolle 2013). Dead Angophoraleaves are light yellow like hay, whereas dead Eucalyp-tus leaves are dark brown with internal decay. Ango-phora has sparse small emergent oil glands with fourpapillate cap cells and bristle glands, unlike Eucalyptuswith internal oil glands (Baker & Smith 1920, Ladiges1984, Brooker & Nicolle 2013). Emergent oil glands in

Angophora costata have densities of 950/cm2 and A.hispida lacks glands, whereas the large internal islandglands of Eucalyptus haemastoma are 1650/cm2, andthose of E. piperita are 350/cm2 (Brooker & Nicolle2013). Angophora and Eucalyptus have cuticles ofabout the same thickness (2–4 μm: Baker & Smith1920, Ladiges 1984). A reasonable hypothesis account-ing for greater decay of Eucalyptus than Angophora isless abundant decay-resistant lignin and leakage ofabundant oil glands, allowing internal pathways for bac-teria additional to substomatal chambers.

Angophora and Eucalyptus have similar loadings ofantibiotic terpenes, including bicyclogermacrene (Dun-lop et al. 1999). The myrtaceous taxa Angophora,Eucalyptus, Melaleuca and Kunzea all have such terpe-nes with antibiotic effects on bacteria (Keszei et al.2010). Furthermore, these taxa are also more denselyveined with lignin than other taxa. An inverse assay ofleaf lignin is nitrogen content, because nitrogen is notfound in refractory lignin, but is abundant in othereasily decomposed tissues (Berg et al. 1996, Berg &Matzner 1997, Berg 2000). The nitrogen content ofleaves of Angophora costata is 3600–5700 ppm, andthat of Eucalyptus haemastoma is ca 5700 ppm, butassociated legume leaves have a nitrogen content of8800–20 500 ppm (Hannon 1956). Melaleuca linariifo-lia and Acacia linifolia also share leaves and phyllodes(respectively) that are narrow with a thick lignin midribthat remains recognizable when the soft tissues aredecayed. These oils and lignin may explain the pre-ferred preservation of these taxa compared with otherspecies (Fig. 5).

Preservation of bracken (Pteridium esculentum) wasexceptional and due to recent death of a clone in onequadrat remaining from a ground fire of five years pre-viously. This thicket of bracken was so dense that fewother leaves shed by surrounding plants found their wayinto that litter sample.

Fig. 7. Rarefaction analysis of specimens of species in the modern litter and living vegetation at Mars Creek, compared with late Permian (Lepi-dopteris callipteroides) and late Early Triassic (Dicroidium zuberi) fossil plant assemblages from the Sydney Basin.

ALCHERINGA LEAF LITTER PRESERVATION 9

Alectryon subcinereus (Sapindaceae) was the onlynonsclerophyll leaf showing preferential preservation inthe leaf litter, albeit with a low preservation index likeKunzea ambigua. This species has flanged papillaepartly occluding abaxial stomata, and also persistent tri-chomes (Pole 2010), which would not be especiallyeffective in blocking bacterial entry on death. Antibioticchemicals are yet not reported from Alectryon sub-cinereus, but are known from other species of thatgenus. Polysaccharide extracts of Alectryon tomentosusare antimicrobial against Bacillus subtilis and cytotoxic(Aboutabi et al. 2016), and Alectryon excelsus has highlevels of hydrocyanic acid (Greshoff 1909). A chemicaldefense in the form of terpenes (Hutchinson & Black-man 2002, Llorens et al. 2009), may also explain thesimilarly puzzling preferential preservation of Nothofa-gus over associated Eucalyptus leaves in some Victorianleaf litters (Steart et al. 2005, 2009).

This study also shows that thickly cuticled, succu-lent, hirsute, pubescent and pinnate leaves, and greenstems are not favored for preservation in leaf litters.Banskia serrata has one of the thickest cuticles (5–6 μm), with abaxial stomata encrypted in grooves (Jor-dan et al. 2008), but low PI. Also under-represented inthe leaf litter were hirsute leaves (Grevillea linearifolia),pubescent leaves (Xanthosia pilosa, X. tridentata, Phyl-lanthus hirtellus), pinnate leaves (Isopogon anethifolius,Lomatia silaifolia), broad phyllodes (Acacia longifolia),terete leaves (Hakea sericea), semisucculent leaves(Dampieria stricta) and photosynthetic stems (Cassythapubescens). Ericaceae and Proteaceae are locallydiverse, but poorly represented in the leaf litter. Con-spicuously absent in the leaf litter compared with stand-ing vegetation immediately above are nonsclerophyllspecies (Callicoma serratifolia, Ceratopetalum apeta-lum, Ligustrum sinense, Lonicera japonica), mostgrasses and low herbs (Centella asiatica). These obser-vations support the hypothesis of Spicer (1991) that par-enchyma, trichomes and stomatal openings providepathways for bacterial invasion (Spicer 1991).

Relevance to paleobotany

These modern litter samples are analogs for certainkinds of fossil plant collections found atop paleosols(Retallack 1977a, 1977b, 1999), or what also have beencalled ‘foliar roofs’ (Krassilov 1975, Holmes 1982),‘obrution deposits’ (Libertín et al. 2009, Stevens & Hil-ton 2009, Dunn et al. 2012), ‘silcrete plant fossils’(Rozefelds 1990, Carpenter et al. 2011), ‘ganisters’(Retallack 1977a, Percival 1983) and ‘nut beds’(Manchester 1994, Retallack et al. 2000). Mars Creekleaf litter has dominance of only a few species of thick,heavily lignified leaves, and thus a broad plateau in therarefaction curve of litter, not seen in the rising rarefac-tion curve of its standing vegetation (Fig. 7). The rar-efaction curves of the Late Permian Lepidopteris

callipteroides and Early Triassic Dicroidium zuberiassemblages of the Sydney Basin are similar to MarsCreek litter, but other fossil floras have a continuallyrising rarefaction curve (Barclay et al. 2003, Wilf et al.2003) more like standing vegetation of Mars Creek(Fig. 7).

For some fossil floras, this kind of dominance is notjust at one locality as expected from local derivation ofleaf litter samples (Burnham 1989, Greenwood 1992),but regional. The late Permian (Changhsinghian) fossilseed fern Lepidopteris callipteroides, for example, isfound throughout the 36 000 km2 Sydney Basin (Mayneet al. 1974) in a low diversity (13 species) assemblageof megafossil plants associated with nutrient-poor pale-osols (Retallack 1999, Retallack et al. 2011). The earlyTriassic (Spathian) fossil seed fern Dicroidium zuberi iscomparably widespread in a megafossil flora of 23 spe-cies associated with infertile quartzose paleosols (Retal-lack 1977a, 1977b, 1997b). An indication of greatersource floral diversity comes from 77 species of dis-persed pollen and spores in the Changhsinghian assem-blage, and 43 species in the Spathian assemblage(Retallack 1995). Collection bias may also explain thesedifferences, because these Permian and Triassic assem-blages have not been collected as extensively as Middleand Late Triassic fossil floras (Anderson et al. 1996,Holmes & Anderson 2013). Both Lepidopteris andDicroidium leaves are remarkable for the thickness oftheir cuticles and durability during maceration of inter-nal organic matter in nitric acid: a week or more ofdigestion is needed to clear the leaves to the extent tak-ing only 20 min for the same genera in Middle Triassicfossil floras (Retallack 1999, Holmes & Anderson2013). This maceration resistance is found throughoutthe Sydney Basin for both floral assemblages (Retallacket al. 2011), so is not due to changing coal rank, whichincreases from the southern to northern Sydney Basin(Diessel 1992). Both Lepidopteris and Dicroidium arealso much larger (up to 40 cm long) and more dissectedthan eucalypt leaves, and migrated southward into theSydney Basin with paleotemperature rise (Retallack2013b). These are both common fossils in their assem-blages and so taken as zonal indicators, but like modernAngophora in this study, they may not have been sodominant in their original vegetation. What appear tohave been low diversity fossil floras, may originallyhave been more diverse like the associated palynoflora(Retallack 1995) and the remarkable extant flora grow-ing on the Hawkesbury Sandstone (Beadle 1968, Hop-per 2009).

Both fossil zones also have some localities withlocally abundant ferns, like the clump of Pteridiumesculentum seen in Mars Creek: local dominance of thefern Cladophlebis carnei was seen in both the Lepi-dopteris callipteroides Zone (Retallack 1999) andDicroidium zuberi Zone (Retallack 1977a, 1977b).These fossil ferns may have been preserved despite thincuticles and lack of lignification by local abundance

10 G. J. RETALLACK ALCHERINGA

and rapid burial of leaf litters. Such rarities, conflationof leaf litter and lacustrine assemblages, and evolution-ary diversification account for high diversity of MiddleTriassic fossil plants (Anderson et al. 1996, Holmes andAnderson 2013).

Paleosols associated with Dicroidium zuberi in theNewport Formation and Lepidopteris callipteroides inthe Coal Cliff Sandstone (Table 1) are noncalcareousand low in nutrient cations and phosphorus (Retallack1977a, 1977b, 1997a, 1999, Retallack et al. 2011), likemodern soils on the Hawkesbury Sandstone (Hannon1956, Beadle 1962, 1968, Chapman & Murphy 1989,McKenzie et al. 2004). The Avalon and Warriewoodpaleosols associated with Dicroidium zuberi (Retallack,1997a) and the Wybung paleosol associated with Lepi-dopteris callipteroides (Retallack, 1999) are clay poorwith chemical index of alteration [Al2O3/(Al2O3+CaO+-Na2O+K2O) × 100] of 80–95%, and 200–700 ppmP2O5. This compares well with 230–720 ppm P2O5 forlocal soils on the Narrabeen Group and 23–263 ppm forsoils on the Hawkesbury Sandstone (Beadle 1962). Thepaleosols also compare well with the chemical index ofalteration of modern soils on the Hawkesbury Sand-stone of 70–94% (Chittleborough 1991).

This deep weathering during the Early Triassic wasdue to quartz-rich parent material but also may berelated to CO2 greenhouse crises revealed by the stom-atal index of Lepidopteris and carbon isotopic composi-tion of organic matter (Retallack et al. 2011, Retallack2013b). Paleosols reveal that CO2 greenhouse spikescoincide with spikes of mean annual precipitation andmean annual temperature, which increased chemicalweathering at these particular fossiliferous levels (Retal-lack et al. 2011, Retallack 2013b). These were times(latest Changhsinghian and late Spathian) of the mostsevere life crises in the history of life (Retallack 1999),and the marked peinomorphic sclerophylly of Dicroid-ium zuberi and Lepidopteris callipteroides may in partbe related to these exceptional atmospheric crises.Dicroidium zuberi and other species of Lepidopteris athigher stratigraphic levels have had much less sclero-phyllous leaves (Holmes & Anderson 2013, Retallack2013b).

The rarefaction curve for the Mars Creek leaf litterhas a broad plateau (Fig. 7) unlike many fossil plantassemblages from lakes (Barclay et al. 2003, Wilf et al.2003, Ellis & Johnson 2013), which continue to riselike the rarefaction curve of extant species found nearMars Creek (Fig. 7). Lakes are better preservationalenvironments for plants than leaf litters of acidic soils(Drake & Burrows 1980, Ferguson 1985, Gastaldoet al. 1989, Spicer 1991), or fertile volcaniclastic pale-osols (Retallack et al. 2000, Retallack & Dilcher 2012).Little trace of fossil plants is left by many other pale-osols, such as former grassland soils (Mollisols: Retal-lack 2013c) and tropical forest soils (Oxisols: Retallack& German-Heins 1994). The 29 species recovered from

litter compared with 74 species in Mars Creek, or 34%of the flora represented in litter is comparable with thehigh end of 5–35% of species in other studies of leaflitters (Burnham 1989, Greenwood 1992, Ellis & John-son 2013). This agreement of total species representa-tion is surprising considering the very differenttechnique of tree survey used in those other studies, andthe eutrophic soils of those forests. What stands out inthe Mars Creek litter and comparable fossil floras(Retallack 1977a, 1977b, 1999) is the overwhelmingdominance of only a few kinds of leaves.

The various features of leaves found to promotepreferential preservation in this study are also apparentin the fossil record. High vein densities can be mea-sured in both cleared and impression leaves (Retallacket al. 2011). Thickness of cuticle, and features, such astrichomes, oil glands, and stomatal occlusion can beobserved in suitably preserved fossil plants (Mösleet al. 1998, Retallack et al. 2011). There is also the pro-spect of biochemical characterization of fossil leaves(Niklas et al. 1978). Although cutan content of cuticlehas been linked to preferential preservation (Tegelaaret al. 1991), subsequent studies have challenged thatview (Gupta et al. 2006). More promising from thefindings of this study are analyses of polysaccharideand phenolic antibiotic compounds in fossils (Mösleet al. 1998, Yang et al. 2005).

ConclusionsThe Eucalyptus woodland litter samples for this studyare an under-representation of the actual floral diversity(only 29 out of 74 species), comparable with other sam-pled leaf litters (Burnham 1989, Greenwood 1992, Ellis& Johnson 2013). However, unlike those other studiesof nonsclerophyll and eutrophic leaf litters, the MarsCreek oligotrophic eucalypt litters show a very strongbias toward sclerophyll leaves of Angophora costata,Melaleuca linearifolia and Eucalyptus haemastoma, inthat order. These sclerophyll genera are also prominenttrees in the living vegetation, but in the reverse order ofimportance. These strong biases in the leaf litter can beattributed to extreme sclerophylly in the most com-monly preserved leaves (Retallack 2009), and these fea-tures are considered peinomorphic adaptations to verylow soil fertility in phosphorus, nitrogen and potassium(Hannon 1956, Beadle 1968). Key adaptations of Ango-phora costata and Melaleuca linearifolia are high lignincontent with dense venation. The sparse emergent oilglands of Angophora costata also provide fewer entrypoints for bacteria than internal oil glands of Eucalyptushaemastoma and Melaleuca linearifolia (Ladiges,1984). All these myrtaceous taxa Angophora, Eucalyp-tus, Melaleuca and Kunzea also have oils with preserva-tive terpenes (Dunlop et al. 1999, Keszei et al. 2010).Thick cuticles and sunken stomata are ineffectivedefenses against decay, because leaves rot from the

ALCHERINGA LEAF LITTER PRESERVATION 11

inside out by bacteria entering stomatal and trichomebases (Spicer 1991, Tian et al. 1997).

The modern sclerophyll leaf litter of Mars Creek isnot similar to other leaf litters studied for comparisonwith fossil assemblages (Burnham 1989, Greenwood1992, Ellis & Johnson 2013), and is a better analog forfossil floras of low-nutrient terrains (Retallack 1977a,1977b, 1999). Fossil floras with overwhelming domi-nance of only a few sclerophyll leaf types may havebeen derived from a much more diverse original flora,and represent an extremely biased record of past vegeta-tion. Examples of such fossil floras associated withnutrient-poor paleosols in the Sydney Basin include thelatest Permian (Changhsinghian) flora dominated byLepidopteris callipteroides (Retallack 1999, Retallacket al. 2011), and late Early Triassic (Spathian) Dicroid-ium zuberi flora (Retallack 1977a, 1977b, 1997b).

AcknowledgmentsI thank Frank J. Burrows for inspiration to conduct thisstudy, and access to scales and other equipment.Heather Adamson, Alison Edgecombe, Jean Vanry andN.C.W. Beadle helped with plant identification. BobSpicer offered useful comments on an earlier draft.

Disclosure statementNo potential conflict of interest was reported by the author.

ReferencesABOUTABI, E.A., ELSAWI, S.A., RASHED, K.N. & RAGAB, N.A., 2016.

Composition of polysaccharides from Alectryon tomentosus F.Muell. leaves with potential cytotoxic and antimicrobial activities.Starch 68, 929–934.

AERTS, R., 1997. Climate, leaf litter chemistry and leaf litter decompo-sition in terrestrial ecosystems: a triangular relationship. Oikos 79,439–449.

ANDERSON, J.M., ANDERSON, H.M., FATTI, L.P. & SICHEL, H., 1996. TheTriassic explosion (?): a statistical model for extrapolating biodi-versity based on the terrestrial Molteno Formation. Paleobiology22, 318–328.

ASTORGA, G.A., JORDAN, G.J. & BRODRIBB, T., 2016. Towards under-standing the fossil record better: Insights from recently depositedplant macrofossils in a sclerophyll-dominated subalpine environ-ment. Review of Palaeobotany and Palynology 233, 1–11.

Australian Bureau of Meteorology, 2016. Climate statistics for Aus-tralian locations, Macquarie Park 066156. http://www.bom.gov.au/climate/averages/tables/ accessed 13.12.2016

BAKER, R.T. & SMITH, H.G., 1920. A Research on Eucalypts and theirEssential Oils. N.S.W. Government Printer, Sydney, 295 pp.

BARCLAY, R.S., JOHNSON, K.R., BETTERTON, W.J. & DILCHER, D.L.,2003. Stratigraphy and megaflora of a KT boundary section inthe eastern Denver Basin, Colorado. Rocky Mountain Geology 38,45–71.

BARRATT, B.C., 1968. Micromorphological observations on the effectsof land use differences on some New Zealand soils. New ZealandJournal of Agricultural Research 11, 101–130.

BEADLE, N.C.W., 1962. Soil phosphate and the delimitation on plantcommunities in eastern Australia II. Ecology 43, 281–288.

BEADLE, N.C.W., 1968. Some aspects of the ecology and physiologyof Australian xeromorphic plants. Australian Journal of Science30, 348–355.

BEADLE, N.C.W., 1981. The Vegetation of Australia. CambridgeUniversity Press, Cambridge, 690 pp.

BERG, B., 2000. Litter decomposition and organic matter turnover innorthern forest soils. Forest Ecology and Management 133, 13–22.

BERG, B., JOHANSSON, M.B., EKBOHM, G., MCCLAUGHERTY, C., RUTI-

GLIANO, F. & SANTO, A.V.D., 1996. Maximum decomposition lim-its of forest litter types: a synthesis. Canadian Journal of Botany74, 659–672.

BERG, B. & MATZNER, E., 1997. Effect of N deposition on decomposi-tion of plant litter and soil organic matter in forest systems. Envi-ronmental Reviews 5, 1–25.

BLACKBURN, W.M. & PETR, T., 1979. Forest litter decomposition andbenthos in a mountain stream in Victoria, Australia. Archiv fürHydrobiologie 86, 453–498.

BROOKER, M.I.H. & NICOLLE, D., 2013. Atlas of Leaf Venation and OilGland Patterns in the Eucalypts. C.S.I.R.O Press, Melbourne,232 pp.

BURNHAM, R.J., 1989. Relationships between standing vegetation andleaf litter in a paratropical forest: implications for paleobotany.Review of Palaeobotany and Palynology 58, 5–32.

CARPENTER, R.J., GOODWIN, M.P., HILL, R.S. & KANOLD, K., 2011. Sil-crete plant fossils from Lightning Ridge, New South Wales: newevidence for climate change and monsoon elements in the Aus-tralian Cenozoic. Australian Journal of Botany 59, 399–425.

CHAPMAN, G.A. & MURPHY, C.L., 1989. Soil Landscapes of Sydney 1/100,000 Sheet. Soil Conservation Service of New South Wales,Sydney, 1 pp.

CHITTLEBOROUGH, D.J., 1991. Indices of weathering for soils andpalaeosols formed on silicate rocks. Australian Journal of EarthSciences 38, 115–120.

CURTIS, J.J. & MACINTOSH, R.P., 1951. An upland forest continuum inthe prairies frontier border region of Wisconsin. Ecology 32, 476–496.

DIESSEL, C., 1992. Coal-Bearing Depositional Systems. Springer, Ber-lin, 721 pp.

DRAKE, H. & BURROWS, C.J., 1980. The influx of potential macrofos-sils into Lady Lake, north Westland, New Zealand. New ZealandJournal of Botany 18, 257–274.

DUNLOP, P.J., BIGNELL, C.M., BROOKER, M.I.H., BROPHY, J.J. & HIB-

BERT, D.B., 1999. Use of gas chromatograms of essential leaf oilsto compare eight taxa of genus Angophora (Myrtaceae): possiblerelationships to the genus Eucalyptus. Biochemical and SystematicEcology 27, 815–830.

DUNN, M.T., ATKINSON, P., LACEFIELD, J. & RISCHBIETER, M., 2012.Winslowia tuscumbiana gen. et sp. nov. (Chaloneriaceae): a cor-mose, heterosporous, ligulate lycopsid reconstructed from theinside out from the Pride Mountain Formation (Late Mississip-pian/Serpukhovian) of Northern Alabama. International Journalof Plant Sciences 173, 96–111.

ELLIS, B. & JOHNSON, K.R., 2013. Comparison of leaf samples frommapped tropical and temperate forests: Implications for interpreta-tions of the diversity of fossil assemblages. Palaios 28, 163–177.

FERGUSON, D.K., 1985. The origin of leaf-assemblages–new light onan old problem. Review of Palaeobotany and Palynology 46,117–188.

FOX, B.J., FOX, M.D. & MCKAY, G.M., 1979. Litter accumulation afterfire in a eucalypt forest. Australian Journal of Botany 27, 157–165.

GASTALDO, R.A., 1989. Preliminary observations on phytotaphonomicassemblages in a subtropical/temperate Holocene bayhead delta:Mobile Delta, Gulf Coastal Plain, Alabama. Review of Palaeob-otany and Palynology 58, 61–83.

GASTALDO, R.A., ADENDORFF, R., BAMFORD, M., LABANDEIRA, C.C.,NEVELING, J. & SIMS, H., 2005. Taphonomic trends of macrofloralassemblages across the Permian–Triassic boundary, Karoo Basin,South Africa. Palaios 20, 479–497.

GASTALDO, R.A., BEARCE, S.C., DEGGES, C.W., HUNT, R.J., PEEBLES,M.W. & VIOLETTE, D.L., 1989. Biostratinomy of a Holoceneoxbow lake: a backswamp to mid-channel transect. Review ofPalaeobotany and Palynology 58, 47–59.

GASTALDO, R.A., PFEFFERKORN, H.A. & DIMICHELE, W.A., 1995. Tapho-nomic and sedimentologic characterization of roof-shale floras. In

12 G. J. RETALLACK ALCHERINGA

Historical Perspectives of Early Twentieth Century CarboniferousPaleobotany in North America (W.H. Darrah Volume). LYONS,P.C., MOREY, E.D. & WAGNER, R.H., eds, Geological Society ofAmerica Memoir 185, 341–352

GREENWOOD, D.R., 1992. Taphonomic constraints on foliar physiog-nomic interpretations of Late Cretaceous and Tertiary palaeocli-mates. Review of Palaeobotany and Palynology 71, 149–190.

GRESHOFF, M., 1909. Phytochemical investigations at Kew. RoyalBotanical Garden Kew Bulletin Miscellaneous Information 10,397–418.

GUPTA, N.S., COLLINSON, M.E., BRIGGS, D.E., EVERSHED, R.P. & PAN-COST, R.D., 2006. Reinvestigation of the occurrence of cutan inplants: implications for the leaf fossil record. Paleobiology 32,432–449.

HANNON, N.J., 1956. The status of nitrogen in Hawkesbury Sandstonesoils and their plant communities in the Sydney District. I. Lin-nean Society of New South Wales Proceedings 81, 119–143.

HANNON, N.J., 1958. The status of nitrogen in Hawkesbury Sandstonesoils and their plant communities in the Sydney District. II. Lin-nean Society of New South Wales Proceedings 83, 65–84.

HILL, R.S. & GIBSON, N., 1986. Distribution of potential macrofossilsin Lake Dobson, Tasmania. Journal of Ecology 74, 373–384.

HOLMES, W.B.K., 1982. The Middle Triassic flora from Benolong, nearDubbo, central-western New South Wales. Alcheringa 6, 1–33.

HOLMES, W.B.K. & ANDERSON, H.M., 2013. A synthesis of the richGondwana Triassic megafossil flora from Nymboida, Australia. InThe Triassic System. TANNER, L.H., SPIELMANN, J.A. & LUCAS,S.G., eds, New Mexico Museum of Natural History and ScienceBulletin 61, 298–305.

HOPPER, S.D., 2009. OCBIL theory: towards an integrated understand-ing of the evolution, ecology and conservation of biodiversity onold, climatically buffered, infertile landscapes. Plant and Soil322, 49–86.

HOWARTH, R.W. & FISHER, S.G., 1976. Carbon, nitrogen, and phospho-rus dynamics during leaf decay in nutrient-enriched streammicroecosystems. Freshwater Biology 6, 221–228.

HUTCHINSON, J.P. & BLACKMAN, A.J., 2002. Biogenic volatile organiccompounds in the ambient air and plant emissions of a Tasmaniannative forest, and the relationship between emission rates andessential oil content. Royal Society of Tasmania Papers and Pro-ceedings 136, 173–183.

JORDAN, G.J., WESTON, P.H., CARPENTER, R.J., DILLON, R.A. & BRO-

DRIBB, T.J., 2008. The evolutionary relations of sunken, covered,and encrypted stomata to dry habitats in Proteaceae. AmericanJournal of Botany 95, 521–530.

KANADE, R., TADWALKAR, M., KUSHALAPPA, C. & PATWARDHAN, A.,2008. Vegetation composition and woody species diversity atChandoli National Park, northern Western Ghats, India. CurrentScience 95, 637–646.

KESZEI, A., BRUBAKER, C.L., CARTER, R., KÖLLNER, T., DEGENHARDT, J.& FOLEY, W.J., 2010. Functional and evolutionary relationshipsbetween terpene synthases from Australian Myrtaceae. Phyto-chemistry 71, 844–852.

KRASSILOV, V.A., 1975. Paleoecology of Terrestrial Plants: Basic Prin-ciples and Techniques. Translated from Russian by H. HARDIN,Wiley, New York, NY, 284 pp.

LADIGES, P.Y., 1984. A comparative study of trichomes in AngophoraCav. and Eucalyptus L’Herit.-a question of homology. AustralianJournal of Botany 32, 561–574.

LAKE, J.C. & LEISHMAN, M.R., 2004. Invasion success of exotic plantsin natural ecosystems: the role of disturbance, plant attributes andfreedom from herbivores. Biological Conservation 117, 215–226.

LAVELLE, P., BLANCHART, E., MARTIN, A., MARTIN, S. & SPAIN, A.,1993. A hierarchical model for decomposition in terrestrialecosystems: application to soils of the humid tropics. Biotropica25, 130–150.

LIBERTÍN, M., OPLUŠTIL, S., PŠENIČKA, J., BEK, J., SÝKOROVÁ, I. &DAŠKOVÁ, J., 2009. Middle Pennsylvanian pioneer plant assem-blage buried in situ by volcanic ash-fall, central Bohemia, CzechRepublic. Review of Palaeobotany and Palynology 155, 204–233.

LLORENS, L., LLUSIÀ, J., MURCHIE, E.H., PEÑUELAS, J. & BEERLING, D.J.,2009. Monoterpene emissions and photoinhibition of ‘living fos-sil’ trees grown under CO2 enrichment in a simulated Cretaceous

polar environment. Journal of Geophysical Research: Biogeo-sciences 114, G01005. doi:10.1029/2008JG000802.

MANCHESTER, S.R., 1994. Fruits and seeds of the Middle Eocene nutbeds flora, Clarno Formation, Oregon. Paleontographica Ameri-cana 58, 1–20.

MAYNE, S.J., NICHOLAS, E., BIGG-WITHER, A.L., RASIDI, J.S. & RAINE,M.J., 1974. Geology of the Sydney Basin—a review. Bureau ofMineral Resources, Geology and Geophysics, Australia, Bulletin149, 1–229.

MCKENZIE, N., JACQUIER, D., ISBELL, R. & BROWN, K., 2004. Aus-tralian Soils and Landscapes: An Illustrated Compendium.CSIRO Press, Melbourne, 432 pp.

MÉNDEZ-TORIBIO, M., ZERMEÑO-HERNÁNDEZ, I. & IBARRA-MANRÍQUEZ,G., 2014. Effect of land use on the structure and diversity ofriparian vegetation in the Duero River watershed in Michoacán,Mexico. Plant Ecology 215, 285–296.

MITCHELL, P.B. & HUMPHREYS, G.S., 1987. Litter dams and microter-races formed on hillslopes subject to rainwash in the SydneyBasin, Australia. Geoderma 39, 331–357.

MÖSLE, B., COLLINSON, M.E., FINCH, P., STANKIEWICZ, B.A., SCOTT,A.C. & WILSON, R., 1998. Factors influencing the preservation ofplant cuticles: a comparison of morphology and chemical compo-sition of modern and fossil examples. Organic Geochemistry 29,1369–1380.

NIKLAS, K.J., BROWN, R.M., SANTOS, R. & VIAN, B., 1978. Ultrastruc-ture and cytochemistry of Miocene angiosperm leaf tissues. U.S.National Academy of Science Proceedings 75, 3263–3267.

PERCIVAL, C.J., 1983. The Firestone Sill ganister, Namurian, northernEngland—the A2 horizon of a podzol or podzolic palaeosol. Sedi-mentary Geology 36, 41–49.

POLE, M., 2010. Cuticle morphology of Australasian Sapindaceae.Botanical Journal Linnean Society 164, 264–292.

PONGE, J.F., 2003. Humus forms in terrestrial ecosystems: a frameworkto biodiversity. Soil Biology and Biochemistry 35, 935–945.

RETALLACK, G.J., 1977a. Triassic palaeosols in the upper NarrabeenGroup of New South Wales. Part I: Features of the palaeosols.Geological Society of Australia Journal 23, 383–399.

RETALLACK, G.J., 1977b. Reconstructing Triassic vegetation of easternAustralasia: a new approach for the biostratigraphy of Gondwana-land. Alcheringa 1, 253–283.

RETALLACK, G.J., 1981. Comment on ‘Reinterpretation of the deposi-tional environment of the Yellowstone “fossil forests”’ by W.J.Fritz. Geology 9, 52–53.

RETALLACK, G.J., 1985. Triassic fossil plant fragments from shallowmarine rocks of the Murihiku Supergroup, New Zealand. RoyalSociety of New Zealand Journal 15, 1–26.

RETALLACK, G.J., 1995. Permian–Triassic life crisis on land. Science267, 77–80.

RETALLACK, G.J., 1997a. Dinosaurs and dirt. In Dinofest International.WOLBERG, D.L., STUMP, E. & ROSENBERG, G.D., eds, Academy ofNatural Sciences, Philadelphia, PA, 345–359.

RETALLACK, G.J., 1997b. Palaeosols in the upper Narrabeen Group ofNew South Wales as evidence of Early Triassic palaeoenviron-ments without exact modern analogues. Australian Journal ofEarth Science 44, 185–201.

RETALLACK, G.J., 1999. Post-apocalyptic greenhouse paleoclimaterevealed by earliest Triassic paleosols in the Sydney Basin, Aus-tralia. Geological Society of America Bulletin 111, 52–70.

RETALLACK, G.J., 2009. Comment on ‘Hot, wet, cold or toxic? Revisit-ing the ecological significance of leaf cuticular micromorphology’by M. Haworth and J. McElwain. Palaeogeography Palaeoclima-tology Palaeoecology 273, 207–208.

RETALLACK, G.J., 2013a. A short history and long future of paleope-dology. In New frontiers in Paleopedology and Terrestrial Paleo-climatology. DRIESE, S.G. & NORDT, L.E., eds, Society ofEconomic Paleontologists and Mineralogists Special Paper 44,5–16.

RETALLACK, G.J., 2013b. Permian and Triassic greenhouse crises.Gondwana Research 24, 90–103.

RETALLACK, G.J., 2013c. Global cooling by grasslands in the geologi-cal past and near future. Annual Review of Earth and PlanetarySciences 41, 69–86.

ALCHERINGA LEAF LITTER PRESERVATION 13

RETALLACK, G.J., BESTLAND, E.A. & FREMD, T.J., 2000. Eocene andOligocene paleosols and environmental change in central Oregon.Geological Society of America Special Paper 344, 1–192.

RETALLACK, G.J. & DILCHER, D.L., 1988. Reconstructions of selectedseed ferns. Annals of the Missouri Botanical Garden 75, 1010–1057.

RETALLACK, G.J. & DILCHER, D.L., 2012. Core and geophysical logsversus outcrop for interpretation of Cretaceous paleosols in theDakota Formation of Kansas. Palaeogeography Palaeoclimatol-ogy Paleoecology 329–330, 47–63.

RETALLACK, G.J. & GERMAN-HEINS, J., 1994. Evidence from paleosolsfor the geological antiquity of rain forest. Science 265, 499–502.

RETALLACK, G.J., SHELDON, N.D., CARR, P.F., FANNING, M., THOMPSON,C.A., WILLIAMS, M.L., JONES, B.G. & HUTTON, A., 2011. MultipleEarly Triassic greenhouse crises impeded recovery from Late Per-mian mass extinction. Palaeogeography PalaeoclimatologyPalaeoecology 308, 233–251.

RODRIGUES, R.R., MARTINS, S.V. & DE BARROS, L.C., 2004. Tropicalrain forest regeneration in an area degraded by mining in MatoGrosso State, Brazil. Forest Ecology and Management 190, 323–333.

ROZEFELDS, A.C., 1990. A mid-Tertiary fossil rainforest from Capellacentral Queensland. In International Organization for Paleob-otany Conference Proceedings. DOUGLAS, J.G. & CHRISTOPHEL,D.C., eds, A-Z Printers, Melbourne, 123–136.

SANBORN, P. & PAWLUK, S., 1989. Microstructure diversity in Ah hori-zons of black chernozemic soils, Alberta and British Columbia(Canada). Geoderma 45, 221–240.

SCOTT, A.C., 1977a. Sedimentological and ecological control of West-phalian B plant assemblages from West Yorkshire. Yorkshire Geo-logical Society Proceedings 41, 461–508.

SCOTT, A.C., 1977b. A review of the ecology of upper Carboniferousplant assemblages, with new data from Strathclyde. Palaeontol-ogy 20, 447–473.

SCOTT, A.C. & COLLINSON, M., 1983. Investigating fossil plant beds.Geology Teaching 7, 114–122.

SCOTT, A.C., MATTEY, D.P. & HOWARD, R., 1996. New data on the for-mation of Carboniferous coal balls. Review of Palaeobotany andPalynology 93, 317–331.

SILVEIRA, F.A., NEGREIROS, D., BARBOSA, N.P., BUISSON, E., CARMO,F.F., CARSTENSEN, D.W., CONCEIÇÃO, A.A., CORNELISSEN, T.G.,ECHTERNACHT, L., FERNANDES, G.W. & GARCIA, Q.S., 2016. Ecol-ogy and evolution of plant diversity in the endangered camporupestre. Plant and Soil 403, 129–152.

SNIDERMAN, J.M.K., JORDAN, G.J. & COWLING, R.M., 2013. Fossil evi-dence for a hyperdiverse sclerophyll flora under a non-Mediter-ranean-type climate. National Academy of Sciences United StatesProceedings 110, 3423–3428.

SPICER, R., 1991. Plant taphonomic processes. In Taphonomy: Releas-ing the Data Locked in the Fossil Record. ALLISON, P.A. &BRIGGS, D.E.G., eds, Topics in Geobiology 9, 71–113.

STEART, D.C., BOON, P.I. & GREENWOOD, D.R., 2006. Overland trans-port of leaves in two forest types in southern Victoria, Australiaand its implications for palaeobotanical studies. Royal Society ofVictoria Proceedings 118, 65–74.

STEART, D.C., BOON, P.I., GREENWOOD, D.R. & DIAMOND, N.T., 2002.Transport of leaf litter in upland streams of Eucalyptus andNothofagus forests in south-eastern Australia. Archiv für Hydrobi-ologie 156, 43–61.

STEART, D.C., GREENWOOD, D.R. & BOON, P.I., 2005. Paleoecologicalimplications of differential biomass and litter production incanopy trees in Australian Nothofagus and Eucalyptus forests.Palaios 20, 452–462.

STEART, D.C., GREENWOOD, D.R. & BOON, P.I., 2009. The chemicalconstraints upon leaf decay rates: taphonomic implications amongleaf species in Australian terrestrial and aquatic environments.Review of Palaeobotany and Palynology 157, 358–374.

STEVENS, L.G. & HILTON, J., 2009. Ontogeny and ecology of the fili-calean fern Oligocarpia gothanii (Gleicheniaceae) from the Mid-dle Permian of China. American Journal of Botany 96, 475–486.

TEGELAAR, E.W., KERP, H., SCHENCK, P.A. & DE LEEUW, J.W., 1991.Bias of the paleobotanical record as a consequence of variationsin the chemical composition of higher vascular plant cuticles.Paleobiology 17, 133–144.

TIAN, X.J., TAKEDA, H. & ANDO, T., 1997. Application of a rapid thinsection method for observations on decomposing litter in morhumus form in a subalpine coniferous forest. Ecological Research12, 289–300.

WILF, P., CÚNEO, N.R., JOHNSON, K.R., HICKS, J.F., WING, S.L. &OBRADOVICH, J.D., 2003. High plant diversity in Eocene SouthAmerica: evidence from Patagonia. Science 300, 122–125.

WOLFE, J.A., 1993. A method of obtaining climatic parameters fromleaf assemblages. U.S. Geological Survey Bulletin 2040, 1–71.

YANG, H., HUANG, Y., LENG, Q., LEPAGE, B.A. & WILLIAMS, C.J., 2005.Biomolecular preservation of Tertiary Metasequoia fossil lager-stätten revealed by comparative pyrolysis analysis. Review ofPalaeobotany and Palynology 134, 237–256.

14 G. J. RETALLACK ALCHERINGA